AgMoDB
ModelsAgentsOrchestratorsEvalsCompositesVisualizeIndustry
AgMoDB by @mistakeknot
All Models

Kimi K2.5 (Reasoning)

January 27, 2026Last synced Nov 22, 2026, 8:17 AM262K context
Effort
  • No reasoning
  • Reasoning

Blended Price

$1.20/M

Input Price

$0.60/M

Output Price

$3.00/M

Speed

—

TTFT

—

Benchmark Scores

Intelligence Index
—
Coding Index
—
Math Index
—
MMLU Pro
— / 100
GPQA Diamond
87.9 / 100
HLE
30.7 / 100
LiveCodeBench
— / 100
SciCode
— / 100
MATH-500
— / 100
AIME
— / 30
AIME 2025
— / 30
IFBench
70.2 / 100
LCR
78.0 / 100
Terminal-Bench Hard
34.8 / 100
τ²-Bench
95.9 / 100

Performance Radar

Kimi K2.5 (Reasoning)
Field Average

Hidden, no data: Output Speed

External Benchmarks

AA-Omniscience Indexaa_omniscience
-8.1 / 100
ARC-AGI-1arc_agi
65.3 / 100
ARC-AGI-1 Cost per Taskarc_agi
0.1
ARC-AGI-2arc_agi
11.8 / 100
ARC-AGI-2 Cost per Taskarc_agi
0.3
ARC-AGI-1benchmark_matrix
65.3 / 100
ARC-AGI-2benchmark_matrix
35.0 / 100
AIME 2024benchmark_matrix
96.1
AIME 2025 (Matrix)benchmark_matrix
96.0
BrowseCompbenchmark_matrix
74.9
BRUMO 2025benchmark_matrix
98.3
CMIMC 2025benchmark_matrix
91.3
Codeforces Ratingbenchmark_matrix
2350.0
GPQA Diamond (Matrix)benchmark_matrix
87.6
HLE (Matrix)benchmark_matrix
50.2
HMMT Feb 2025benchmark_matrix
95.4
HMMT Nov 2025benchmark_matrix
89.2
HumanEvalbenchmark_matrix
95.0
IFEvalbenchmark_matrix
90.0
LiveCodeBench (Matrix)benchmark_matrix
85.0
MATH-500 (Matrix)benchmark_matrix
98.0
MathArena Apex 2025benchmark_matrix
8.8
MathVisionbenchmark_matrix
84.2
MMLUbenchmark_matrix
92.0
MMLU-Pro (Matrix)benchmark_matrix
87.1
MMMU-Probenchmark_matrix
78.5
OSWorld (Unversioned Matrix Import)benchmark_matrix
62.0 / 100
SimpleQAbenchmark_matrix
45.0
SMT 2025benchmark_matrix
90.6
Terminal-Bench 2.0benchmark_matrix
43.2
Video-MMUbenchmark_matrix
86.6
browsecompbenchmark_matrix
74.9
FrontierMathbenchmark_matrix
28.0 / 100
MMMUbenchmark_matrix
84.0 / 100
simpleqabenchmark_matrix
45.0
SWE-bench Verifiedbenchmark_matrix
76.8 / 100
AA-Omniscience AccuracyPredicted
43.4 / 100
AA-Omniscience Hallucination RatePredicted
95.4 / 100
Aider PolyglotPredicted
82.3 / 100
AIMEPredicted
0.4 / 30
AIME 2025Predicted
0.9 / 30
AlpacaEval 2.0 LCPredicted
36.4 / 100
AlpacaEval 2.0 RawPredicted
26.7 / 100
BFCL (Berkeley Function Calling)Predicted
54.9
BigCodeBench CompletePredicted
45.6 / 100
BigCodeBench InstructPredicted
36.2 / 100
AA Intelligence Index (Matrix)Predicted
63.0
AA Long Context Reasoning (Matrix)Predicted
75.2
Arena-Hard AutoPredicted
88.9
CritPtPredicted
5.6
GSM8KPredicted
95.9
IFBench (Matrix)Predicted
61.7
IMO 2025Predicted
38.7
MRCR v2Predicted
77.9
SWE-bench ProPredicted
47.2
Tau-Bench Telecom (Matrix)Predicted
96.2
Terminal-Bench 1.0Predicted
39.8
USAMO 2025Predicted
26.1
Aider PolyglotPredicted
0.2
BALROGPredicted
0.0
BIG-Bench HardPredicted
3.0
BoolQPredicted
0.6
CAD-EvalPredicted
2.7
Chess PuzzlesPredicted
0.3
CyBenchPredicted
0.6
DeepResearchBenchPredicted
0.5
GdpvalPredicted
0.4
GeoBenchPredicted
0.0
GSM8K (Epoch)Predicted
2.4
GSOPredicted
0.0
HellaSwagPredicted
0.0
Lech Mazur WritingPredicted
8.6
METR Time HorizonsPredicted
40.8
OTIS Mock AIME 2024–2025Predicted
1.0
PIQAPredicted
0.8
PosttrainbenchPredicted
0.0
SimpleQA Verified (Epoch)Predicted
0.5
The Agent CompanyPredicted
23.9
TriviaQAPredicted
31.9
VPCTPredicted
0.6
WinoGrandePredicted
1.4
GAIA Level 1Predicted
46.3
GAIA Level 2Predicted
32.1
GAIA Level 3Predicted
33.1
GAIAPredicted
34.1 / 100
LegalBenchPredicted
83.6 / 100
LiveCodeBenchPredicted
0.8 / 100
LongBench v2 EasyPredicted
55.6
LongBench v2 HardPredicted
40.4
LongBench v2Predicted
41.5 / 100
MATH-500Predicted
0.9 / 100
MathVistaPredicted
70.0 / 100
MLE-bench (Original)Predicted
60.1 / 100
MMLU ProPredicted
0.9 / 100
MMTU Table UnderstandingPredicted
56.6 / 100
MT-BenchPredicted
7.8 / 10
NoLiMa (NIAH)Predicted
89.9 / 100
OCRBench v2Predicted
86.3 / 100
RE-BenchPredicted
98.0
SWE-bench LitePredicted
79.5 / 100
tau-bench RetailPredicted
88.0 / 100
WebArenaPredicted
0.7 / 100
WildBenchPredicted
41.3
browsecompbrowsecomp
74.9
BullshitBenchbullshitbench
52.0 / 100
Arena ELO: Codingchatbot_arena
1436.0
Arena ELO: Creative Writingchatbot_arena
1415.0
arena_elo_documentchatbot_arena
1431.0
Chatbot Arena ELOchatbot_arena
1445.0
arena_elo_visionchatbot_arena
1269.0
Ale Benchepoch_ai
821.6
Apex Agentsepoch_ai
2.5
Arc Agi 2epoch_ai
0.1
Cl Benchepoch_ai
0.2
Cl Bench Lifeepoch_ai
0.1
Critptepoch_ai
0.0
Cursorbenchepoch_ai
0.3
Eci Scoresepoch_ai
149.3
Enigma Evalepoch_ai
0.0
Epoch Capabilities Indexepoch_ai
148.0
FictionLiveBenchepoch_ai
0.9
Frontiercodeepoch_ai
0.0
Frontiersweepoch_ai
15.7
Hleepoch_ai
0.2
OSWorld (Epoch Import, Version Unverified)epoch_ai
63.3 / 100
Posttrainbenchepoch_ai
0.1
Processed Data For Eciepoch_ai
0.1
Scicodeepoch_ai
0.5
SimpleQA Verified (Epoch)epoch_ai
0.3
Vending Bench 2epoch_ai
1198.5
WebDev Arenaepoch_ai
1436.4
EQ-Bench 3eqbench
1303.0
HuggingFace Downloads (30d)hf-downloads
406888.0
HuggingFace Likeshf-downloads
37.0
LiveBench Codinglivebench
77.9 / 100
LiveBench Data Analysislivebench
61.4 / 100
LiveBench Languagelivebench
77.7 / 100
LiveBench Mathlivebench
84.9 / 100
LiveBench Overalllivebench
69.1 / 100
LiveBench Reasoninglivebench
76.0 / 100
AIME 2026matharena
95.8
MedQA (USMLE)medqa
94.4
OmniDocBench Overallomnidocbench
1.0 / 100
Open LLM Averageopen_llm_leaderboard
14.6 / 100
Open LLM: BBHopen_llm_leaderboard
49.7 / 100
Open LLM: GPQAopen_llm_leaderboard
27.7 / 100
Open LLM: IFEvalopen_llm_leaderboard
22.5 / 100
Open LLM: MATH Level 5open_llm_leaderboard
2.7 / 100
Open LLM: MMLU-PROopen_llm_leaderboard
30.0 / 100
Open LLM: MUSRopen_llm_leaderboard
39.8 / 100
OpenRouter Token Volumeopenrouter_rankings
302000000000.0
OpenRouter Weekly Trendopenrouter_rankings
18.0
SimpleBenchsimplebench
46.8 / 100
SWE-bench Verifiedswebench
70.8 / 100
Vectara Factual Consistencyvectara_hallucination
85.8 / 100
Vectara Hallucination Ratevectara_hallucination
14.2 / 100
WeirdMLweirdml
45.6 / 100
SWE-rebenchswe_rebench

vcontinuous-leaderboard · tasks-created:2026-01-01:2026-03-01 · Kimi K2.5 · harness swe-rebench-react@tools · 5 attempts · pass@1 · swe-rebench:standardized:128k-or-model-limit · provider Kimi K2.5__tools · record Kimi K2.5__tools:1767225600000:1772323200000

View run source

Source metadata
{
  "passAt5": 60.952380952380956,
  "developer": "moonshot",
  "instanceType": "model",
  "contextPolicy": "128k unless the model supports less",
  "attemptsPerTask": 5,
  "totalTokenUsage": 1361950.016043956,
  "modelReleaseDate": "2026-01-26",
  "scoreStandardError": 0.7979212617639928,
  "cachedTokenPercentage": 92.2213727004562,
  "modelReleaseTimestamp": 1769385600000,
  "standardContextTokens": 128000,
  "averageInstanceCostUsd": 0.19543760866117216,
  "potentialContamination": true,
  "aggregateTaskRangeTimestamp": {
    "to": 1772323200000,
    "from": 1767225600000
  },
  "publishedTaskRangeTimestamp": {
    "to": 1772323200000,
    "from": 1767225600000
  }
}

Evidence observed 2026-03-01

49.0 / 100

How Kimi K2.5 (Reasoning) Compares

Axes

Filters

$0.00 – $67.5
≥ 1.4

Providers

ai2ai21-labsalibabaanthropicarceeawsazurebaidu
openai
zai
meta
nvidia
xai
upstage
deepseek
alibaba
mistral
nous-research
aws
google
anthropic
xiaomi
stepfun
kimi
reflection
minimax
azure
inclusionai
reka-ai
cohere
bytedance_seed
inception
tencent
thinking-machines
Bubble size = Context Window
8191.00
2000000.00
Compare with other models